"All or None" Approach in Precision Oncology Sometimes Poses a Dilemma

Maurie Markman, MD

During the past several years, there has been considerable discussion—and even intense debate— within the oncology literature regarding the currently documented benefits of precision cancer medicine. Some have questioned the realistic potential of whether this approach will have a meaningful impact within the realm of cancer care in the relatively near term.

Certainly, few would disagree that a major potential benefit of the precision medicine process is the ability to increase the likelihood that a given antineoplastic strategy will achieve a favorable effect in a specific individual with cancer while minimizing the chances that a clinically inactive, very expensive, and possibly quite toxic drug will be employed. The existing experience with mutant KRAS in metastatic colon cancer is an excellent example, one might even suggest a “poster child,” of a highly clinically useful molecular marker that achieves this goal. Extensive retrospective reviews of prospectively conducted evidence-based clinical trials have revealed that metastatic colon cancers possessing a KRAS mutation very rarely respond (as measured by standard RECIST criteria) to the administration of monoclonal antibody anti-EGFR therapy.¹

In fact, in several studies the reported response rate to cetuximab was zero in metastatic colon cancer with a documented KRAS mutation.¹ Conversely, in the presence of wild-type KRAS there is a reasonable opportunity for an objective response to be observed. In this setting, several studies have revealed the response rate to cetuximab to be in the range of 30% to 40%.¹

Similarly, few would argue today with the utility of testing for the presence of wild-type EGFR in predicting the essential absence of meaningful clinical activity of a small-molecule tyrosine kinase inhibitor of EGFR when employed as second-line therapy for metastatic non—small cell lung cancer (NSCLC).² For example, in a randomized trial comparing single-agent erlotinib with docetaxel in this clinical setting, an objective response rate of only 3% was found among 100 evaluable patients receiving erlotinib.²

Some Benefit Without Biomarker

However, as we continue to explore the potential for molecular tumor markers to help define both the patient populations likely, or most unlikely, to respond to a particular targeted antineoplastic agent, it is critical to understand that not all—or perhaps even most—determinations of the absence of clinical utility will be as simple and straightforward as is the case with mutant KRAS in metastatic colon cancer or wild-type EGFR in NSCLC.Consider the following examples. In a provocative report of the clinical activity of osimertinib (AZD9291), a novel EGFR inhibitor for the treatment of patients with EGFR inhibitor-resistant NSCLC, the investigators found an impressive objective response rate of 61% among 127 patients whose cancers were shown to contain an EGFR T790M mutation, an abnormality previously demonstrated to be responsible for a substantial percentage of resistance to routinely employed first-generation tyrosine kinase inhibitors of EGFR.³ Of note, specific eligibility for entry into this trial required “radiologically documented disease progression after previous treatment with EGFR tyrosine kinase inhibitors.”³ The authors appropriately concluded that “AZD9291 was highly active in patients with lung cancer with the EGFR T790M mutation” after progression on prior anti-EGFR therapy.³

However, what was not stated in the conclusion of this paper, and possibly not elsewhere when the activity of this drug is further discussed, is the observation that among the 61 patients who had progressed on prior anti-EGFR therapy but where an EGFR T790M could not be identified, the objective response rate was 21%.³ This percentage is certainly lower than that observed in the presence of the specific mutation, as was the median progression-free survival of 2.8 months versus 9.6 months for patients with the mutation. But would it be objectively reasonable to conclude, or even suggest, that this agent is biologically and clinically inactive in the absence of the EGFR T790M mutation?

In mid-November, the FDA approved osimertinib for patients with advanced NSCLC who have progressed after prior anti-EGFR therapy and whose tumors harbor the EGFR T790M mutation. Yet is it truly appropriate to deny payment for the administration of this drug to a patient in that NSCLC setting in the absence of an EGFR T790M mutation? Yes, nearly two of three patients treated with the drug appear to exhibit an objective response if this mutation is present. However, even in the absence of this specific mutation approximately one in five treated individuals may be able to achieve such a response.

A similar statement can be made for the potential use of PD-L1 expression as a biomarker to assist in the selection of the patient population most likely to benefit from the delivery of novel PD-1/ PD-L1 inhibitor antibody therapy. For example, in a trial comparing nivolumab with dacarbazine in previously untreated patients with metastatic melanoma, 53% of individuals classified as demonstrating positive PD-L1 levels achieved an objective response compared with a response rate of only 33% in individuals with a cancer that was either PD-L—or negative indeterminate.⁴ Notably, the objective response rate was higher than observed in the control arm in both cases.

Again, while the opportunity for a response to this agent was clearly greater (20% higher in this trial) in the presence of this biomarker, even in its absence one of three patients achieved a response and, likely, meaningful clinical benefit.

Therefore, in the absence of a very clear and validated signal that unequivocally identifies a particular molecular biomarker as strongly predictive for the lack of clinical utility of a specific targeted antineoplastic strategy, caution is advised in excluding patients from receiving the potential benefit of a drug documented to be associated with a highly favorable clinical outcome in a given disease setting.

Maurie Markman, MD, editor-in-chief, is president of Medicine & Science at Cancer Treatment Centers of America, and clinical professor of Medicine, Drexel University College of Medicine. maurie.markman@ctca-hope.com.

References

1. Allegra CJ, Jessup JM, Somerfield MR, et al. American Society of Clinical Oncology provisional clinical opinion: testing for KRAS gene mutations in patients with metastatic colorectal carcinoma to predict response to anti-epidermal growth factor receptor monoclonal antibody therapy. J Clin Oncol. 2009; 27(12):2091-2096.
2. Garassino MC, Martelli O, Broggini M, et al. Erlotinib versus docetaxel as second-line treatment of patients with advanced non—small-cell lung cancer and wild-type EGFR tumors (TAILOR): a randomised controlled trial. Lancet Oncol. 2013;14(10):981-988.
3. Jänne PA, Yang JC, Kim DW, et al. AZD9291 in EGFR inhibitor—resistant non–small-cell lung cancer. N Engl J Med. 2015;372(18):1689-1699.
4. Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated melanoma without BRAF mutation. N Engl J Med. 2015;372(4):320-330.